FWIW, I am installing a system where the drainback tank is pumped through the floor, I used oxygen-barrier PEX and will be using a plastic tank and a bronze circulator.

gg

Yes the PEX is the same : 100 meters in the floor, and 66 meters from tank to collector and back...

http://certipubli.cstb.fr/2016/138/2935-153-2113.pdf

Not got the pumps sorted yet, but they will be Salmson becos that's what my oppo has got and he told me to go to the same route !

http://www.salmson.com/fileadmin/user_upload/UPLOAD/product/notice_technique/SXS_NT_FR_50Hz.pdf

My tank is supposedly good to 90°C, I doubt it will get anywhere near that ! (especially in December)

gg

To keep people happy why not put in a heat exchanger some where to separate the solar thermal from the UFH.
Options
Tube heat exchanger in the heat store to feed UFH
Plate heat exchanger external to the heat store to feed UFH
Tube heat exchanger in the heat store to to take the output from the solar thermal.

If you use the tube heat exchanger in the heat store to to take the output from the solar thermal then water volume will be fairly small so it would be economical to use antifreeze/inhibitor in the solar thermal and not have a drain back system.

Posted By: Herodotusit's going to be too much of a PITA to arrange a consistent fall across the whole array back to the store

Just a thought - do you know about the Tichelmann system ?

https://uk.grundfos.com/service-support/encyclopedia-search/tichelmann-system.html


Maybe watch this first !
https://www.youtube.com/watch?v=VlAfTYTA0Z0

gg

@gyrogear. Yeah, you've reminded me why I didn't want to use glycol in the first place. We're going to have 6 panels for a total nominal output of 10.8kw. Our plumber says he doesn't recommend using drainback with this number of panels.

Does anyone know if there's a practical limit on how many panels you can have in series and still have drainback working effectively? Do we need to do two sets of 3 in parallel or will 6 drain ok in series? They're the flat plate style panels. It's also complicated because we're running entirely on PV electric, and there's a ~4.5m height difference between the store and the top of the panels - which entails quite a powerful pump, particularly if we put all the panels in series (higher frictional losses). Oh, and we're at 1200m of altitude, so the underpressure thing is potentially going to mean our panels boil at a lower temperature as well . Although I really like the idea of drainback in principle (simple is good) I wonder if in practice we're just going to encounter a bunch of other, less obvious complexity. See https://backend.orbit.dtu.dk/ws/files/140025523/drain.pdf for example.

@Peter - I have been thinking seriously about this option. There are a couple of practical challenges/ downsides with it:

1) I had hoped to avoid the glycol closed solar circuit. Apart from the intrinsic efficiency downsides of glycol + coil, and the additional components required, it's going to create a bunch of problems in summer. The system is being designed to provide a decent proportion of our space-heating requirements during the winter; during the summer when we have more sun and no space heating needs, it will be massively oversized. A pressurized glycol system is going to give us a bunch of headaches here. We'd probably need to think about splitting the system into two parts and draining one of them, or having some sort of heatsink to dissipate the excess in summer - neither of which I'm very excited about. One possible compromise here is an indirect drainback system with a coil. It's not ideal because most of the efficiency benefits of drainback systems come from losing the heat exchanger component, but at least it avoids the overheating issues.

2) Stores with 3 coils - two supply and one DHW exchanger - are not that common. Those that exist pretty much all, AFAICT, have one configuration: the DHW coil runs the full height up the middle, then there's a solar coil at the bottom and a smaller, secondary coil towards the upper middle of the store. The primary use case for this is to have an additional, higher temperature heat source input - e.g. a boiler. Running the UFH through this coil isn't ideal because you're skipping the whole of the bottom half of the store - which is exactly where you want to be working for a low-temperature circuit like UFH. This is certainly still better than returning overly hot water to the bottom of the store as would most likely happen with a PHX on the UFH circuit, but it's still not great.

Posted By: HerodotusDoes anyone know if there's a practical limit on how many panels you can have in series and still have drainback working effectively? Do we need to do two sets of 3 in parallel or will 6 drain ok in series?

The answers to those questions are in here :

https://www.youtube.com/watch?v=CW9YVaPW9wU

"collector arrangements" at cue 12':49''

especially important = linear expansion on long systems : ends at 15:41...

relatively explicit

gg

Posted By: HerodotusOne possible compromise here is an indirect drainback system with a coil. It's not ideal because most of the efficiency benefits of drainback systems come from losing the heat exchanger component, but at least it avoids the overheating issues.

Sounds like this would be the way to go

Herodotus,

can I ask why you have a 20kW log boiler? At 35kWh/day heat loss in winter, you're looking at 2kW power input. 20kW seems a lot. Maybe you inherited it with the house, or want to do a single batch burn to charge the TS once, and then forget about the boiler? Even so, about 2 hours running would fully load the TS from totally cold (looks like 750 litre on drawing?).

In terms of solar thermal, I always install drainback, to avoid a pressurised system, with summer boiling, winter freezing, pressurised charging with glycol, and the typical conditions in UK of sun/cloud/sun/cloud, whereby drainback holds the heated water inside the thermal envelope, not out on the roof. Not suggesting pressurised is wrong, or worse, just my opinion as the person the client will call if there's ever an issue.

6 panels in series will give a very high output temp, even in winter. Like you, I'm wondering about 2 triples or even 3 pairs in parallel? If you have a variable speed pump, the solar controller would vary the pump speed / flow rate to suit.

Whilst I mention solar controller, I assume you will use your solar controller to manage the boiler pump?

Your separate (duplicate) DHW cyl worries me a bit, in terms of legionella control. The 100 L could sit unused at uncontrolled temps for periods of time. Would you consider NOT taking DHW from the thermal store? Use it as a buffer tank to receive the boiler and solar input. Then draw UFH and a hot feed to the DHW cyl? 100 litres seems a bit small anyway.

I think a better answer, (maybe?), install two solar coils, one top and one bottom, with a solar transfer valve. That will direct hottest solar input to top, with cooler towards the base. That way you'll make better use of high temp solar input, having better DHW reservoir at the top of the TS. NoGets rid a separate DHW cyl, a couple of3 port valves, and some control logic to decide when to draw DHW from which cyl.

So essentially as per your drwg, but remove the DHW cyl and the two 3-ports, and add a second solar coil above the lower one, plus a solar transfer valve.

The high lift to the solar panels - could you just put 2 cheap domestic circ pumps in series, avoiding going to a much more expensive "commerical" high pressure pump?

Hope there's something amongst that of use to you?

I personally don't buy a "system". It's only 4 or 5 components, so why pay someone else to group those in a box?

- panel capable of draining (no that many which the manuf actually state as drainback, but some do state it, and others are drain back but manuf doesn't state it. Does the water run out under gravity in your chosen orientation?). Look for suppliers of drain back complete systems, then buy that panel, or find the manufacturer of that panel. Get the manuf internal pipe work/connection drwg,and confirm for yourself that it's drainable.

- circ pump

- drainback tank (get this made by the therm store manufacturer Ne**rk Cop Cyl). It's only a little insulated pot with a few ports, one with a dip leg

- solar controller (take your pick. I choose ones that also allow me to run a log boiler stove, and have 3 temp sensors for the TS)

Slightly necro-ing my own thread... but hey. I'm still puzzling this out and really running out of time to make decision on this now.

Posted By: GreenPaddycan I ask why you have a 20kW log boiler?

Two reasons: firstly because for dedicated boilers (as opposed to a LBS with a back boiler) it's actually quite hard to find anything under 15-20kw. I don't want anything that burns wood in the house because it takes up loads of space and it makes mess/dust. Secondly, as you surmised, because I'd rather be able to load it once a day, do a fast, hot, efficient burn and dump that heat into the TS. i don't want to be popping out to the boiler room all day to keep it stoked up.

I take the point about legionella. The tricky thing here is balancing having a guaranteed supply of DHW with avoiding legionella problems. Since we don't have an automatic heat source that is guaranteed to produce > 60 degree temperatures there's no perfect solution to this problem. The safest option, as you say, is to dispense with the secondary hot water tank, and this was always my preference. The downside is that if we have a really cold night and the floor sucks up a lot of heat, we could wake up in the morning with little/no hot water. I guess some extra intelligence in the control programme could help to avoid that (I'm building my own controller from a Raspberry PI + a bunch of 1-wire thermo sensors). Maybe the best fix for this is just a cheap gas backup for emergencies.

The health of the floor is still concerning to me. It's the one thing that is utterly inaccessible and almost impossible to fix if it goes wrong, so it makes sense to take all possible precautions against it becoming clogged up with crap. I know there are different schools of thought on this but there do seem to be a lot of plumbers (and most of the manufacturers/suppliers) who strongly recommend having a sealed system for the floor. I'm wavering between: Peter's suggestion of running the floor through a coil and leaving the rest of the system unpressurized; or alternatively keeping just the solar circuit unpressurized with a coil of some sort into the TS. The choice basically comes down to what the log boiler needs. At the moment we're looking at one that wants a pressurized circuit so I think the second option is probably better.

If you use the TS (or even a separate TS) for DHW then you don't have to worry about legionella.

If you use a coil in the bottom half of the TS to supply a closed-circuit UFH system then you'll answer most of your rust etc worry and a filter will fix the rest. A coil in the bottom shouldn't take heat from the top of the store, as long as it stays stratified.

So if you use a coil in the top half of the store for DHW you should be OK ordinarily. An immersion halfway up the store will provide backup for DHW, or alternatively an inline heater could be used.

Coil(s) for the solar input as well. Then the TS could be unpressurised with an open-vented stove or could be pressurised with your proposed model.

I say bottom half and top half, but sizing the volumes appropriately may mean some different ratio. Or you could have two separate TS in parallel, which some seem to prefer.

Posted By: djhIf you use the TS (or even a separate TS) for DHW then you don't have to worry about legionella.

If you use a coil in the bottom half of the TS to supply a closed-circuit UFH system then you'll answer most of your rust etc worry and a filter will fix the rest. A coil in the bottom shouldn't take heat from the top of the store, as long as it stays stratified.

So if you use a coil in the top half of the store for DHW you should be OK ordinarily. An immersion halfway up the store will provide backup for DHW, or alternatively an inline heater could be used.

Coil(s) for the solar input as well. Then the TS could be unpressurised with an open-vented stove or could be pressurised with your proposed model.

I say bottom half and top half, but sizing the volumes appropriately may mean some different ratio. Or you could have two separate TS in parallel, which some seem to prefer.

This is more or less what I'm leaning towards. I'm not going to have two TS in parallel, although this is probably ideal, because (1) money and (2) space.

I'm looking at something like this: https://biomasalamanca.es/depositos-hygienico-2-serpentin/291-tanque-hygienico-con-serpentin-de-acero-inox-800-litros2-apoyo.html

It has a coil at the bottom for solar input, one in the middle that I could use to extract heat for the floor, and then a third SS coil with a preheat at the bottom and main coil at the top. The only thing I don't like is that it's uncoated steel (which is surprisingly common here). But I guess with appropriate inhibitor that's not an issue, particularly if I go with a closed pressurized circuit for the boiler.

The main downside with this approach is that there are two coil heat exchangers between the panels and the floor. That's a lot of inefficiency, particularly as nothing here gets made of copper If the boiler is on a closed pressurized circuit then in theory there's nothing stopping me pulling the floor water directly from the store itself.

Or there's the really radical option, where I buy an 800l PP rain water tank, immerse a copper coil for DHW water production and another one to pull out floor heat, and then run the boiler and the solar drainback directly from the water in the tank. I'll need to find a different boiler that will operate at atmospheric pressure though. And I'm probably opening myself to a world of potential boiler failure nonsense there as well since the drainback circuit of necessity has a lot of oxygen in it. That and the sketch factor of building it all myself from scratch.

Just for info.. We have an open vented thermal store and unpressurized UFH. Same water circulates through the boiler. Dosed the system with Fernox F1 corrosion inhibitor. No worries. Just make sure you put enough in. Don't just tip a bottle in, calculate the number of bottles needed based on the volume and recommend concentration. You might need 3-6 bottles if you have a big thermal store. I was advised not to drain and refill too frequently as that introduces oxygen.

I don't know why, but I get the impression there are people here who are fighting shy of pressurized systems. All you need is an expansion vessel 10% the size of your total capacity, pressure relief valves and a 8 bar meter and you're away.
I have a 48kw gasification boiler feeding a 3000L TS with a coil for DHW take off (not yet connected) and a Solar coil (in the process of ng now)
This is directly fed from the boiler via a Laddomat which opens at something like 62C and progressively bleeds back less and less of the hot to the boiler return as the temp rises. The output from the TS is mixed down with sufficient of the returning cold (cool) to deliver at about 68C

I have a mix of rads and UFH (3 discrete manifolds the largest of which has 3 separate zones) each of these manifolds has a thermostatic mixer valve and pump so the temp of the UFH circuits are reliably constant - vital in my opinion if the TS is being fired from a non-automatic source like a maunually filled gasification boiler. The components are 3.5bar rated, and I run it at 0.5bar (cold) and even when the TS is fully charged to about 75-85C the system doesn't climb much higher than 1.5 to just under 2 bar with an expansion tank of 250L.
The system has been up and running for about 10 years now and I have recently flushed the crud out of biggest (and lowest) UFH manifold in order to find why it was getting less and less efficient. Jet black, of course, but not a vast amount of mush in the actual loops. Turned out the problem was a pair of flexibles connecting the manifold were crudded up completely (this was a solution to a first fix/second fix/trying to plumb with a house full of other trades chasing me/scenario and was never meant to be permanent! A change to 22mm CU fixed that!)

There is nothing to be gained by under-sizing a Gasification boiler/TS combo (apart from the cost). There's something very satisfying about popping out to the barn to check on the TS thermometers on a mild day, only to discover you don't have to do a firing at all today, and on the other hand, a system only *just* big enough for the heat demand, requiring loading/firing 2 or 3 times a day in very cold weather, is a daily routine which gets old very quickly!

The solar panels will be 4 x 18 evac. tubes (again pressurized), obviously a separate sealed system feeding the coil in the TS, switchable to a swimming pool heat exchanger, and to a smaller 300L TS for a new DHW system pre-heat coil. At the moment the switching will be manual, but future-proofing all this with mororized valves!
I'm looking to this addition to affect the 'shoulders' of the heating system: there is perhaps a 2 or 3 weeks in the Spring and Autumn when a full burn isn't necessary every day, but the o/night temps are still low. Added to which I have long grieved at the waste of free heat every time I stand by this south-facing wall barn wall and (even in January after a sunny day) I can feel the heat being radiated off it.
It will be interesting to see just how much contribution 12 sqm of panels makes when I get it up and running.

As my old English teacher, Henry Thomas, used to say, "our policy, like Fords, is one of continuous improvement" !